The organization and development of neuronal systems are problems of great importance in modern biology. This proposal concerns a molecular and genetic analysis of a chemosensory system and is designed with a view to addressing the fundamental problems of (i) the mechanisms of chemosensory transduction, (ii) the principles of chemosensory coding, and (iii) the means by which genes specify the development of a complex neuronal system. Drosophila is an excellent organism in which to address these problems, as it possesses a highly sophisticated olfactory system which may be investigated by powerful genetic and molecular means. Among a set of mutants defective in chemosensory behavior are several which respond abnormally to one chemical but which appear normal in response to other chemicals. Such mutants define genes which may play roles in chemically-specific olfactory pathways; these genes could encode olfactory receptor molecules or other products required for transduction or processing of specific types of olfactory information. A genetic and molecular analysis of one such gene, defined by the 3D18 mutation, is proposed. By characterizing the distribution within the olfactory system of products of genes which are specific to individual functional pathways, we anticipate that we may gain an enhanced understanding of the functional organization of the system, which is, in turn, critical to an understanding of how complex olfactory stimuli are encoded into a form interpretable to the brain. Genetic analysis of the olfD locus is also proposed; we aim to determine whether olfD represents a second class of gene, required for response to all chemicals but not in other sensory modalities, such as vision. Other mutants we have isolated are defective both in visual physiology and chemosensory behavior. Relationships between chemosensory and visual transduction will be sought by determining whether the affected genes correlate with previously characterized visual system genes, and by subjecting visual mutants to chemosensory assays. The feasibility of using a direct molecular approach to isolate genes required for olfactory transduction and information processing will be assessed: we will attempt to isolate antennal- specific cDNA clones using a new method of subtractive hybridization.